From 486281e1d8070d9d47494a7b03a022c3d182394f Mon Sep 17 00:00:00 2001
From: Claude <noreply@anthropic.com>
Date: Thu, 12 Mar 2026 22:08:28 +0000
Subject: [PATCH] Add from_file, from_bytes, and from_directory methods to
 BaseParser

Extends the interop engine's parser base class with convenience
methods for loading data from files, bytes, and directories.
All methods delegate to the existing from_string() method.

Closes #122
---
 healthchain/interop/parsers/base.py | 54 +++++++++++++++++++++++++++++
 1 file changed, 54 insertions(+)

diff --git a/healthchain/interop/parsers/base.py b/healthchain/interop/parsers/base.py
index f97b226f..615d6eb0 100644
--- a/healthchain/interop/parsers/base.py
+++ b/healthchain/interop/parsers/base.py
@@ -1,4 +1,6 @@
 from abc import ABC, abstractmethod
+from pathlib import Path
+from typing import List, Union
 
 from healthchain.interop.config_manager import InteropConfigManager
 
@@ -23,3 +25,55 @@ def from_string(self, data: str) -> dict:
             A dictionary containing the parsed data structure
         """
         pass
+
+    def from_bytes(self, data: bytes, encoding: str = "utf-8") -> dict:
+        """Parse input data from bytes.
+
+        Args:
+            data: The input data as bytes
+            encoding: Character encoding to use when decoding bytes
+
+        Returns:
+            A dictionary containing the parsed data structure
+        """
+        return self.from_string(data.decode(encoding))
+
+    def from_file(self, file_path: Union[str, Path]) -> dict:
+        """Parse input data from a file.
+
+        Args:
+            file_path: Path to the file to parse
+
+        Returns:
+            A dictionary containing the parsed data structure
+
+        Raises:
+            FileNotFoundError: If the file does not exist
+        """
+        path = Path(file_path)
+        return self.from_string(path.read_text(encoding="utf-8"))
+
+    def from_directory(
+        self, directory_path: Union[str, Path], pattern: str = "*.xml"
+    ) -> List[dict]:
+        """Parse all matching files in a directory.
+
+        Args:
+            directory_path: Path to the directory containing files to parse
+            pattern: Glob pattern to match files (default: "*.xml")
+
+        Returns:
+            A list of dictionaries, one per parsed file
+
+        Raises:
+            NotADirectoryError: If the path is not a directory
+        """
+        path = Path(directory_path)
+        if not path.is_dir():
+            raise NotADirectoryError(f"Not a directory: {path}")
+
+        results = []
+        for file_path in sorted(path.glob(pattern)):
+            if file_path.is_file():
+                results.append(self.from_file(file_path))
+        return results